ÿ
 Hello, Fausto Carvalho, I'd like you to meet my friend, Harvey...
>> jb: When the user aborts a packet download, O_Qwker flashes it's 
>> jb: 'updating  pointers' message on the sysop screen, and the 
>> lastread jb: pointers are in fact updated because a repack does 
>> not find new jb: msgs. 
 FC> That's not correct. If the Download fails, pointers should not 
 FC> be updated. Using the parameter QWKAREA in your .cfg file you 
 FC> can even allow the QWK packet to be moved to a file area wher 
 FC> the user can get it until you delete it (preferably upon 
 FC> logoff) That may be occurring because your external transfer 
 FC> engine (DSZ, whatever) is not returning the correct error code. 
 FC> Anything non zero will tell O_QWKer that the download was not 
 FC> successful. 
I have no idea why GSZ wasn't working, or what errorlevels it returned.
I'd forgotten about qwkarea. I will add a user-accessible file area to take 
care of any future crashes.
>> jb: Conversely, in 'netmail' or 'local' rather than 'echomail' 
>> areas, jb: sometimes msgs are downloaded several times is 
>> successive scans. jb: The pointers are apparently not updated. 
 FC> There is another .cfg option to force full area scan, 
 FC> disregarding message pointers. It's FULLSCAN AREA. I use for my 
 FC> main local area. O_QWKer has currently this feature hardwired 
 FC> for Netmail area. The rationale is that these areas have 
 FC> frequent message killing and renumbering, so trusting the 
 FC> message pointers could lead to users not receiving messages. 
 FC> Maybe this should be reworked... Note taken! 
For echo areas, AFAIK no one has ever missed a message. When everything 
works, it seems to work very well. 
:-)
jB-)
---
---------------